From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.11]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 2EF8E363C51 for ; Wed, 25 Feb 2026 09:05:17 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=fail smtp.client-ip=192.198.163.11 ARC-Seal:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1772010321; cv=fail; b=NQPPa082T+0sYfDmPcdd0zO2Pc5XpBBrgBmCUtWZ4ftyeBPlI30yNLtLyOOe5XEqCySzdCjfYmh/MxUU0ncWJOQKY377MHb5jCELPxKrit26GG1WqBCAzbdggeFPYN9djsEZK6fSlFATmpaJ7knjGQ0PWZ7uu8t2IMDM7aGoh4U= ARC-Message-Signature:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1772010321; c=relaxed/simple; bh=V86QMGohMeAfJMBe84i++X3FxRaf4V+jdOIY4EVNgPE=; h=Message-ID:Date:Subject:To:CC:References:From:In-Reply-To: Content-Type:MIME-Version; b=qg9MV7PugXL/prx6tXBIL8WhJuwfl/QFHUpfKs9fEJRNBKEk7T0j3QVE/QJicJmXK1/7cmHcRgSeoKPqkV6O6ETr5WeSanZe5T4QJwyCV7NAAs4ScJLHK+GLMfGrGCT3J49zPLMicMHdbFDEjsU1vTVtxIJd8Xx/cO8YDGW/et4= ARC-Authentication-Results:i=2; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=intel.com; spf=pass smtp.mailfrom=intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=KIuyjlI1; arc=fail smtp.client-ip=192.198.163.11 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="KIuyjlI1" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1772010318; x=1803546318; h=message-id:date:subject:to:cc:references:from: in-reply-to:content-transfer-encoding:mime-version; bh=V86QMGohMeAfJMBe84i++X3FxRaf4V+jdOIY4EVNgPE=; b=KIuyjlI1E9YcfyVGR3Lx2yI7YwLyMWE8DwQKclpOFrHLGdMVXMnmWK6k dPimTdpCA2xKHxHRCEBo5phkScKghUP97rqnv7mMO1TDGRQJ9iqaM7ruC U4KpSFlI1ilXL0snvV0RSsRzdA4NjQNpFh0t+VbjfHhGEOfd1/1HRZaLn DRLDDUbzD0ch7Lpa3xQrWpZ4jW9pKUb07bRwkq47vJ2hxSB1m6r2EH77+ dUrh6uxN+NZDsPl8YklD8RJlJtjVJpaXenIpeJKZVSlyBB1At4hlLKMJA 3ymGmb7HanTW5PB9hkKMCQuPEZaGvo5IqxAmNMnTVQt5Y4WhfnprQje51 g==; X-CSE-ConnectionGUID: Z6cjQj/ZQFilzk2bX6P0Bg== X-CSE-MsgGUID: qCMFoCJuScGNhgEHnGRtZw== X-IronPort-AV: E=McAfee;i="6800,10657,11711"; a="83668782" X-IronPort-AV: E=Sophos;i="6.21,310,1763452800"; d="scan'208";a="83668782" Received: from fmviesa001.fm.intel.com ([10.60.135.141]) by fmvoesa105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 25 Feb 2026 01:05:17 -0800 X-CSE-ConnectionGUID: yDGcWAg7RlyrIhn0ksJRoQ== X-CSE-MsgGUID: u+Xbzr7aRrGQrwJOer40gA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.21,310,1763452800"; d="scan'208";a="243981038" Received: from orsmsx903.amr.corp.intel.com ([10.22.229.25]) by fmviesa001.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 25 Feb 2026 01:05:17 -0800 Received: from ORSMSX903.amr.corp.intel.com (10.22.229.25) by ORSMSX903.amr.corp.intel.com (10.22.229.25) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.35; Wed, 25 Feb 2026 01:05:16 -0800 Received: from ORSEDG901.ED.cps.intel.com (10.7.248.11) by ORSMSX903.amr.corp.intel.com (10.22.229.25) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.35 via Frontend Transport; Wed, 25 Feb 2026 01:05:16 -0800 Received: from CH4PR04CU002.outbound.protection.outlook.com (40.107.201.26) by edgegateway.intel.com (134.134.137.111) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.2562.35; Wed, 25 Feb 2026 01:05:16 -0800 ARC-Seal: i=1; a=rsa-sha256; s=arcselector10001; d=microsoft.com; cv=none; b=Hf5gfGenasbXtMnJgeijpkMcTifuzCjiNAlGfI2IPJoLZNkx5s9tNAdiBUznx4QSKBaQ/18xSbqP0hU6ksV3Uebr6+oo3BuripJyv1HoDF202rTNzzUxa2HwZwFNbzU+qF8X35qdP0/LIkMj21DUtK/NX6Ksdpsd0mNm/Jb4TQvQGlitZ6agW+lfp2TUsR/tYan8BGkuT5Uz3rz6odRKArl+AJIWOKS3+uHTf643RMWBUWHCjzrRjLVHnjc/iE9FvwoYuZKHfNrMyK1OCaXP3kVp94Po35X92GF0qoabyrCwApRmRXeaJ+559QVgVfVo3ebmly9ZLsXtLsmDAfezdQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector10001; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=i2yyvsbFIpipAIp9TfB4y7te6CMF1sZwH+AWbxRtrYU=; b=v7ecvbk6v00nXXpFWJ/KvtBjS+GlIYpo6pIOm6zR8RxpnADSflq8FdtkDWQpQCsH8u2ZzVefEWCEmn3v7iLX5KeaPFFl34iElfndqzCCfkmPEiy+7HMErGFWTvfJQIom/pr8G/SucvfAfbK22EdCKA+UOtwQDyNyrCMlJg33601FaiqxdHDvel+nvTJrNxnOYdZalknlmwPnbg4C6IQ2SXPJlyC5ITRsvhx9Oiy1WGB0LQ4PNLl2dPdTBvWq8Fxm/QhktxR0Ed051ykLK3gw3xhOEBRo3WX9G9PcHNbxwgjTEIgmxZn7joSQv4cFwCakzMY0pPkUQ1cv29GqWxTtsQ== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=intel.com; dmarc=pass action=none header.from=intel.com; dkim=pass header.d=intel.com; arc=none Authentication-Results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=intel.com; Received: from DM4PR11MB6020.namprd11.prod.outlook.com (2603:10b6:8:61::19) by SN7PR11MB6701.namprd11.prod.outlook.com (2603:10b6:806:26b::11) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.9632.22; Wed, 25 Feb 2026 09:05:12 +0000 Received: from DM4PR11MB6020.namprd11.prod.outlook.com ([fe80::3058:1480:e4ac:5765]) by DM4PR11MB6020.namprd11.prod.outlook.com ([fe80::3058:1480:e4ac:5765%6]) with mapi id 15.20.9632.017; Wed, 25 Feb 2026 09:05:12 +0000 Message-ID: <30c6c21b-6f03-4930-a40a-fcfa045ac362@intel.com> Date: Wed, 25 Feb 2026 17:05:01 +0800 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v2] sched/topology: Check average distances to remote packages To: Kyle Meyer CC: , , , , , , , , , , , , , , , "Peter Zijlstra" , Aubrey Li References: <20260223170314.GU1395266@noisy.programming.kicks-ass.net> Content-Language: en-US From: "Chen, Yu C" In-Reply-To: Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 8bit X-ClientProxiedBy: TPYP295CA0033.TWNP295.PROD.OUTLOOK.COM (2603:1096:7d0:7::17) To DM4PR11MB6020.namprd11.prod.outlook.com (2603:10b6:8:61::19) Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: DM4PR11MB6020:EE_|SN7PR11MB6701:EE_ X-MS-Office365-Filtering-Correlation-Id: 26250d94-5e97-47f2-9ccd-08de744cfb5a X-LD-Processed: 46c98d88-e344-4ed4-8496-4ed7712e255d,ExtAddr X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0;ARA:13230040|366016|1800799024|7416014|376014; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:255.255.255.255;CTRY:;LANG:en;SCL:1;SRV:;IPV:NLI;SFV:NSPM;H:DM4PR11MB6020.namprd11.prod.outlook.com;PTR:;CAT:NONE;SFS:(13230040)(366016)(1800799024)(7416014)(376014);DIR:OUT;SFP:1101; X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?Tlp0empYZDBjM29yeEhVNHM4cy9PSDZ0aGdCZzltTzNqUm1XU2RNK2NqZXBi?= =?utf-8?B?SUp4UGtqOFR6QkRZOEZxaGFBWFFrZEZxU0Z4dFE3RWJjTUF4aDhxV2hPeXo0?= =?utf-8?B?UmRLemtlWDJzSjM0dC9RNXdWUTZGc3pOcEgvVVdvWElaY2hPdG1XN1p3ZUc1?= =?utf-8?B?eENZU2RGemNRSUxmQzVpZHJ2TEgzc0xWNkg4eUZGWmdCMjBRVjN5WkhoYnpl?= =?utf-8?B?UEREbGNucUtTUW5PTHN4enpSM0RSSGcvdzBocE9KbnlOVnpuUDBCOFhuaVNR?= =?utf-8?B?UHpWQ0kydlZPYm80MjJCNDFoWVJLQnA2L08rbVRZRXd6U1p5dERubmJSa0U3?= =?utf-8?B?dkhGNWNsa0xpbWJrSm0ySCtZQ3RuZGJteFVPUTNaOGQraVhpeGRpcUY4allT?= =?utf-8?B?U2JWOHFtRFkzanNXQVRwaEE4cjFVRW5QbWo0YjRwVStyVGh0T0p6ZzlPS2Q1?= =?utf-8?B?c05jV2h5OFpreUF3d2RJWEVQcjhpNTdaUVRObmI2c1FqeExVNzNleXRrTmo3?= =?utf-8?B?RGZhbHRLeklIbnQ4YTJXY3U4cWdRQjZtOFl2eGxmNkFHcTdZWG5rb0RkSFda?= =?utf-8?B?YU83K2NtdzNiOUpDMkNzeVhPZGFhQXpjS05nWEhmR2orRHhHOWZpMG9iYjBM?= =?utf-8?B?OGw3SmlFMXQ4TS9yUFhCREVZOWh0WERTL29MZVZ5VHB0YUo1UjF0TzhIcVNR?= =?utf-8?B?RlhwM2R3RlJmNDNqMmdYSlV6RjZGWFRPK1lRTERPYXl6Q2tYR1JQS0FmbWVt?= =?utf-8?B?Vk9wTUJsanFXb2M3eGloVXcvYmN0ckhLS25ZSmwwdUNOZ1lrTDhSRDQ2Nzl3?= =?utf-8?B?ZWJDN2NkaHpEYVdlYjNIL0RtUXhuY0RnVzg0MFYwaWdxYldTbFgvU2l5MWps?= =?utf-8?B?clZQYWVFckpjZjIzeE0vUUNhVnpTL1F6Y0dHWlEwa1hXaTgwZHZEdEMrR05D?= =?utf-8?B?ejgzOFUwSy9UTmF2eFl0eXlaMlVFdjlYdjltU3dQRmEzTTc3djMxa21yK05w?= =?utf-8?B?NGdWVm0yQlI5TmFjTURoRHhPZ2dMSnRuS3NXRnh3M1lwTFdGL3hlN1UyY0tF?= =?utf-8?B?NC95OXJIVGhvWXZ1TWdxSWRCZllTQm02bG5qcjByK1BXREhGUSt2b1JMdE5C?= =?utf-8?B?VnpmTWRwZDJhWVdQVDNKTWxvTC8yN3RMWS80RDh0RklBUXJJc2FEcktVRkZO?= =?utf-8?B?NTB5NnQ2a0h0b0REV3NPbDBUVFJIaytkWm1EMVVSS1l0TXc3Y29qTk9JY3M0?= =?utf-8?B?ekIxdTVrY2pidEROTFhBWjRNd2Y5MFp6TTNoYUZhVzUvWmFESVNjQkFjbmhC?= =?utf-8?B?TGhscGFTTHBpNGxubld1T21JWFJCTUJ6NlJYbFlXSWFwSXBFSVphVWVEc2tz?= =?utf-8?B?RXBOTGYrcWh6c21wY0tVMHVsRlZRTzFFN0kvbC9uRDJhUTNQbE1FNEc5b2lC?= =?utf-8?B?elVWK2lNQlE1YUdJa2JNMXFFcXBYL2d0N2RpUHcvTkIzOXd6SzZrTDYwUWhI?= =?utf-8?B?WjV0V2t0ZGZrMmFtUzVCZDhZQWpHaHM5cHNueVp3dEtqbCtCS1BXbEJSWFZi?= =?utf-8?B?VFhpM0Q2dWRGVEZCNUc2dTR5Nk45bTQySjgxayt3REZaN29LaHRjK0tTcnlO?= =?utf-8?B?WEF6ckVlM0czWllTUXFlYUFkM1NFRkl0RGpWWTJaMHAwZkFEUThrQ09NWk9p?= =?utf-8?B?dUNTMEx5RnpzYXJHR3AwN2gzYk5qajdaeVorVjQweHFHaHFralhlY3VwMHJF?= =?utf-8?B?SWZZQ3NFNUlUNlNHSEtZdmJqblI5NStYOUN3Z3Z5a21Obm9TNlRPRitQTzFh?= =?utf-8?B?U0o3c2MwYkxIODlaSU9FNHdtVnQ5emZaN20va1ZzN2k5NitpMFpKWjhiSjVW?= =?utf-8?B?NHFHWXFOUDVMdExXendzNEVOalF0eTJEMkRqMHVaU3E4R3k0VW50K1dPSlhq?= =?utf-8?B?alJYQUhXbUZVMlZJWDk3QjYyR2pxeHNsM3NJWnhJck4zcXZXWldZeVNic3Bm?= =?utf-8?B?MmRGVXg0Z3dCSnZxRFo0TXljekFCdEZnRTVML1BRL0xUOE1MNkhlM282dlJ1?= =?utf-8?B?OWlaNWwzS1FycXNoWkFQTzZXeUUySDZiZ2pKTkNTN3dJR2NwVjZJRy9ybTRD?= =?utf-8?B?ejJlbm5zZkxSYmYxQ0Z3SHd2SlZmZDAyRTVhYmVPVlVLeFJpeFVQVC9MK3Zp?= =?utf-8?B?U1EvaGVzRnVpZGIrdTN6dWRoeDBWcnN3ZWlPcGJQT2RycnAydlEySGZCOUE0?= =?utf-8?B?c0FsSk9iZ0N0U2NhdEltYjFxV25YYTB3WjlpN1FENDNjS293Y3Q0bmV5aC9R?= =?utf-8?B?ZHMvaDZjalNRZVpTRlEwalQ1aEI3a2J0emxySHVXaThMOC9SLzBDZz09?= X-MS-Exchange-CrossTenant-Network-Message-Id: 26250d94-5e97-47f2-9ccd-08de744cfb5a X-MS-Exchange-CrossTenant-AuthSource: DM4PR11MB6020.namprd11.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 25 Feb 2026 09:05:11.9548 (UTC) X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted X-MS-Exchange-CrossTenant-Id: 46c98d88-e344-4ed4-8496-4ed7712e255d X-MS-Exchange-CrossTenant-MailboxType: HOSTED X-MS-Exchange-CrossTenant-UserPrincipalName: 5OGwKntlFAv7h7W2DlmZcO5vJvwFrEfWd2FXXMumr/jp6XVZwCTFLghO2FdU5K3q521io/XxU2S/OWqLALs5+Q== X-MS-Exchange-Transport-CrossTenantHeadersStamped: SN7PR11MB6701 X-OriginatorOrg: intel.com Hi Kyle, On 2/25/2026 9:43 AM, Kyle Meyer wrote: > On Mon, Feb 23, 2026 at 06:03:14PM +0100, Peter Zijlstra wrote: >> On Wed, Feb 04, 2026 at 06:24:26PM -0600, Kyle Meyer wrote: >>> Granite Rapids (GNR) and Clearwater Forest (CWF) average distances to >>> remote packages to fix scheduler domains, see [1] for more information. >>> >>> A warning and backtrace are printed when sub-NUMA clustering (SNC) is >>> enabled and there are more than 2 packages because the average distances >>> to remote packages could be different, skewing the single average remote >>> distance. >> >> But earlier Tim said these systems will not have more than 2 packages. >> So what's what? > > We have Intel customer reference boards with 2, 4, and 8 sockets. > Thanks for the info. We were not previously aware that the stock GNR platform will scale up to 8 sockets. >> So what do these new systems look like? > > Here's an 8 socket (2 chassis) HPE system with SNC enabled: > > node 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 > 0: 10 12 16 16 16 16 18 18 40 40 40 40 40 40 40 40 > 1: 12 10 16 16 16 16 18 18 40 40 40 40 40 40 40 40 > 2: 16 16 10 12 18 18 16 16 40 40 40 40 40 40 40 40 > 3: 16 16 12 10 18 18 16 16 40 40 40 40 40 40 40 40 > 4: 16 16 18 18 10 12 16 16 40 40 40 40 40 40 40 40 > 5: 16 16 18 18 12 10 16 16 40 40 40 40 40 40 40 40 > 6: 18 18 16 16 16 16 10 12 40 40 40 40 40 40 40 40 > 7: 18 18 16 16 16 16 12 10 40 40 40 40 40 40 40 40 > 8: 40 40 40 40 40 40 40 40 10 12 16 16 16 16 18 18 > 9: 40 40 40 40 40 40 40 40 12 10 16 16 16 16 18 18 > 10: 40 40 40 40 40 40 40 40 16 16 10 12 18 18 16 16 > 11: 40 40 40 40 40 40 40 40 16 16 12 10 18 18 16 16 > 12: 40 40 40 40 40 40 40 40 16 16 18 18 10 12 16 16 > 13: 40 40 40 40 40 40 40 40 16 16 18 18 12 10 16 16 > 14: 40 40 40 40 40 40 40 40 18 18 16 16 16 16 10 12 > 15: 40 40 40 40 40 40 40 40 18 18 16 16 16 16 12 10 > > 10 = Same chassis and socket > 12 = Same chassis and socket (SNC) > 16 = Same chassis and adjacent socket > 18 = Same chassis and non-adjacent socket Previously, I thought that REMOTE_DISTANCE represents the distance between two nodes on different sockets, but that does not appear to be the case here. I could not find any definition of “20 or double” in the SLIT section of the ACPI specification. Thus, I assume this value of 20 is an artificial threshold. In my view, checking whether all distances above 20 are identical really depends on the specific platform. For example, if we have a distance value such as: 22 = Same chassis and non-adjacent socket then applying the current patch would trigger a warning regardless. That said, since 20 is an artificial threshold, I have a tentative idea:we could normalize the SLIT distances by sorting the slit_dist values, finding the 75th percentile value, keeping all slit_dist values below the 75th percentile unchanged, and treating all slit_dist values above the 75th percentile as remote - assigning them the average remote distance. This way, we could eliminate the arbitrary value of 20. But that might require rewrite.. and for now it is ok to keep 20. > 40 = Different chassis > > Each processor connects to an ASIC (XNC) that acts as a multiplexer, extending > the UPI interconnect across the entire system. > > We don't experience the scheduler domain issue reported by Tim because our SLIT > provides symmetric distances to remote NUMA nodes, but we trigger the WARN_ONCE > because we exceed 2 packages. > >>> This is unnecessary when the average distances to remote packages are >>> the same. >>> >>> Support single average remote distance on systems with more than 2 >>> packages, preventing unnecessary warnings and backtraces, by checking if >>> average distances to remote packages are the same. >> >> >> [ ... ] >>> + pkg = topology_physical_package_id(cpu); >>> + pkg_total_distance[pkg] += distance; >>> + pkg_nr_remote[pkg]++; >> >> This is broken, physical_package_id is not guaranteed to be dense. > > Thank you, I'll fix this. > It seems that topology_logical_package_id() can work here. thanks, Chenyu